Computational Intelligence Laboratory (CIL)

The mission of CIL is to develop methods, tools, and technology for the design and implementation of learning systems which mimic the learning process of humans, and apply them to real world problems. The goal is to solve complex engineering problems which are difficult to deal with using conventional approaches. Current application areas include traffic control/management in telecommunication networks and learning control methodologies for automotive engines, among others. The emphasis in CIL is on collaboration with researchers and practitioners from academia and industry. Some of the current projects in the laboratory are:

 

Computational Intelligence

Click here

Computational intelligence combines elements of learning, adaptation, evolution and fuzzy logic (rough sets) to create systemss that are, in some sense, intelligent. Our research covers neural networks, fuzzy systems and evolutionary computation, including swarm intelligence, as well as their applications.

Computational Neuroscience

Click here

Intelligent Control and Learning Control

Click here
Click here
Click here
Click here
Book Book Book  | Book

Our current interest in this field is neural network-based approaches for intelligent systems and learning control. In particular, our study is focused on neural network-based adaptive critic designs. Adaptive critic designs are designs that approximate dynamic programming in the general case, i.e., approximate optimal control over time in noisy, nonlinear environments. There are many practical problems that can be formulated as to minimize or maximize a measure of cost. It is well-known that dynamic programming is very useful in solving these problems. However, it is often computationally untenable to run true dynamic programming due to the backward numerical process required for its solutions, i.e., due to the “curse of dimensionality.” Over the years, progress has been made to circumvent the “curse of dimensionality” by building a system, called “critic” to approximate the cost function in dynamic programming. The idea is to approximate dynamic programming solutions by using a function approximation structure such as neural networks to approximate the cost function. Our work includes methodology development and applications of adaptive critic designs to automotive engine control.

Click here for publications

This laboratory contains a cluster of multimedia Pentium PCs. The laboratory is located in 910 SEO.